linux-efi.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Matthew Garrett <mjg59-1xO5oi07KQx4cg9Nei1l7Q@public.gmane.org>
To: Matt Fleming <matt.fleming-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
Cc: Ben Hutchings <ben-/+tVBieCtBitmTQ+vhA3Yw@public.gmane.org>,
	Josh Boyer <jwboyer-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
	"stable-u79uwXL29TY76Z2rM5mHXA@public.gmane.org"
	<stable-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
	"linux-efi-u79uwXL29TY76Z2rM5mHXA@public.gmane.org"
	<linux-efi-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
	Seth Forshee
	<seth.forshee-Z7WLFzj8eWMS+FvcfC7Uqw@public.gmane.org>
Subject: Re: efi: be more paranoid about available space when creating variables
Date: Tue, 26 Mar 2013 03:56:00 +0000	[thread overview]
Message-ID: <20130326035600.GA6209@srcf.ucam.org> (raw)
In-Reply-To: <514E31B0.4030305-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>

Ok, so having thought some more about the actual problem (ie, Samsungs 
go wrong if there's too much used space marked as being active, not 
merely too much used space) I think we really want to be looking for the 
amount of active space rather than the remaining space value that 
QueryVariableInfo() gives us. Something like this (absolutely untested, 
provided here for comment) patch? I'll try rigging something up under 
OVMF to test it. This version certainly seems a little over-naive, since 
it won't handle the case of a variable that's being overwritten with 
something of a different size.

commit 263c2ee36c67dfa6d869304a3b5aef7a14f1ec4e
Author: Matthew Garrett <matthew.garrett-05XSO3Yj/JvQT0dZR+AlfA@public.gmane.org>
Date:   Mon Mar 25 13:40:28 2013 -0400

    efi: Distinguish between "remaining space" and actually used space
    
    EFI implementations distinguish between space that is actively used by a
    variable and space that merely hasn't been garbage collected yet. Space
    that hasn't yet been garbage collected isn't available for use and so isn't
    counted in the remaining_space field returned by QueryVariableInfo().
    
    Combined with commit 68d9298 this can cause problems. Some implementations
    don't garbage collect until the remaining space is smaller than the maximum
    variable size, and as a result check_var_size() will always fail once more
    than 50% of the variable store has been used even if most of that space is
    marked as available for garbage collection. The user is unable to create
    new variables, and deleting variables doesn't increase the remaining space.
    
    The problem that 68d9298 was attempting to avoid was one where certain
    platforms fail if the actively used space is greater than 50% of the
    available storage space. We should be able to calculate that by simply
    summing the size of each available variable and subtracting that from
    the total storage space. With luck this will fix the problem described in
    https://bugzilla.kernel.org/show_bug.cgi?id=55471 without permitting
    damage to occur to the machines 68d9298 was attempting to fix.
    
    Signed-off-by: Matthew Garrett <matthew.garrett-05XSO3Yj/JvQT0dZR+AlfA@public.gmane.org>

diff --git a/drivers/firmware/efivars.c b/drivers/firmware/efivars.c
index 7acafb8..731ac7b 100644
--- a/drivers/firmware/efivars.c
+++ b/drivers/firmware/efivars.c
@@ -436,9 +436,12 @@ static efi_status_t
 check_var_size_locked(struct efivars *efivars, u32 attributes,
 			unsigned long size)
 {
-	u64 storage_size, remaining_size, max_size;
+	u64 storage_size, remaining_size, max_size, active_available;
+	struct efivar_entry *entry;
+	struct efi_variable *var;
 	efi_status_t status;
 	const struct efivar_operations *fops = efivars->ops;
+	unsigned long active_size = 0;
 
 	if (!efivars->ops->query_variable_info)
 		return EFI_UNSUPPORTED;
@@ -449,8 +452,16 @@ check_var_size_locked(struct efivars *efivars, u32 attributes,
 	if (status != EFI_SUCCESS)
 		return status;
 
+	list_for_each_entry(entry, &efivars->list, list) {
+		var = &entry->var;
+		get_var_data_locked(efivars, var);
+		active_size += var->DataSize;
+	}
+
+	active_available = storage_size - active_size;
+
 	if (!storage_size || size > remaining_size || size > max_size ||
-	    (remaining_size - size) < (storage_size / 2))
+	    (active_available - size) < (storage_size / 2))
 		return EFI_OUT_OF_RESOURCES;
 
 	return status;

-- 
Matthew Garrett | mjg59-1xO5oi07KQx4cg9Nei1l7Q@public.gmane.org

  parent reply	other threads:[~2013-03-26  3:56 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <1364004995.3728.76.camel@deadeye.wl.decadent.org.uk>
     [not found] ` <1364010441.3728.82.camel@deadeye.wl.decadent.org.uk>
     [not found]   ` <1364010441.3728.82.camel-nDn/Rdv9kqW9Jme8/bJn5UCKIB8iOfG2tUK59QYPAWc@public.gmane.org>
2013-03-23 20:24     ` efi: be more paranoid about available space when creating variables Fleming, Matt
2013-03-23 20:32       ` Matthew Garrett
     [not found]         ` <1364070731.2553.47.camel-tCUS0Eywp2Pehftex57rkxo58HMYffqeLoYNBG0abjxeoWH0uzbU5w@public.gmane.org>
2013-03-23 22:50           ` Matt Fleming
     [not found]             ` <514E31B0.4030305-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2013-03-26  3:56               ` Matthew Garrett [this message]
2013-03-26  4:31                 ` Lingzhu Xiang
     [not found]                   ` <5151248F.2010303-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2013-03-26  4:34                     ` Ben Hutchings
2013-03-26  4:35                     ` Matthew Garrett
     [not found]                 ` <20130326035600.GA6209-1xO5oi07KQx4cg9Nei1l7Q@public.gmane.org>
2013-03-26  7:40                   ` Lingzhu Xiang
2013-04-01 15:13                     ` [PATCH 1/2] efi: Determine how much space is used by boot services-only variables Matthew Garrett
     [not found]                       ` <1364829240-26475-1-git-send-email-matthew.garrett-05XSO3Yj/JvQT0dZR+AlfA@public.gmane.org>
2013-04-01 15:14                         ` [PATCH 2/2] efi: Distinguish between "remaining space" and actually used space Matthew Garrett
     [not found]                           ` <1364829240-26475-2-git-send-email-matthew.garrett-05XSO3Yj/JvQT0dZR+AlfA@public.gmane.org>
2013-04-01 15:50                             ` Ben Hutchings
2013-04-03 13:11                             ` Matt Fleming
     [not found]                               ` <515C2A86.6070606-HNK1S37rvNbeXh+fF434Mdi2O/JbrIOy@public.gmane.org>
2013-04-03 13:48                                 ` Matthew Garrett
2013-04-03 17:12                                   ` Matt Fleming
     [not found]                                     ` <515C62E6.1070004-HNK1S37rvNbeXh+fF434Mdi2O/JbrIOy@public.gmane.org>
2013-04-03 17:20                                       ` Matthew Garrett
2013-04-01 15:42                         ` [PATCH 1/2] efi: Determine how much space is used by boot services-only variables Ben Hutchings
2013-04-03 13:09                         ` Matt Fleming
     [not found]                           ` <515C29EE.8050008-HNK1S37rvNbeXh+fF434Mdi2O/JbrIOy@public.gmane.org>
2013-04-03 13:42                             ` Matthew Garrett
2013-03-26 10:37                   ` efi: be more paranoid about available space when creating variables Matt Fleming
2013-03-26 15:43                     ` Matthew Garrett
     [not found]                       ` <20130326154359.GA20530-1xO5oi07KQx4cg9Nei1l7Q@public.gmane.org>
2013-03-27  9:09                         ` Matt Fleming
     [not found]                           ` <5152B75B.5080305-HNK1S37rvNbeXh+fF434Mdi2O/JbrIOy@public.gmane.org>
2013-04-01  6:31                             ` Ben Hutchings
2013-03-26 23:33                   ` Seiji Aguchi
     [not found]                     ` <A5ED84D3BB3A384992CBB9C77DEDA4D41AF79528-ohthHghroY0jroPwUH3sq+6wyyQG6/Uh@public.gmane.org>
2013-03-27  9:10                       ` Matt Fleming

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20130326035600.GA6209@srcf.ucam.org \
    --to=mjg59-1xo5oi07kqx4cg9nei1l7q@public.gmane.org \
    --cc=ben-/+tVBieCtBitmTQ+vhA3Yw@public.gmane.org \
    --cc=jwboyer-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
    --cc=linux-efi-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=matt.fleming-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org \
    --cc=seth.forshee-Z7WLFzj8eWMS+FvcfC7Uqw@public.gmane.org \
    --cc=stable-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).